Zoological Journal of the Linnean Society, 2011. With 1 figure Phylogenetic signal in the evolution of body colour and spicule skeleton in calcareous sponges ANDRÉ LINHARES ROSSI,1 CLAUDIA AUGUSTA DE MORAES RUSSO,2 ANTONIO MATEO SOLÉ-CAVA,2 HANS TORE RAPP3,4 and MICHELLE KLAUTAU1* 1Universidade Federal do Rio de Janeiro, Instituto de Biologia, Departamento de Zoologia, 21.941-902, Rio de Janeiro, Brazil 2Universidade Federal do Rio de Janeiro, Instituto de Biologia, Departamento de Genética, 21.941-902, Rio de Janeiro, Brazil 3University of Bergen, Department of Biology, N-5020, Bergen, Norway 4University of Bergen, Centre for Geobiology, N-5020, Bergen, Norway Received 16 November 2010; revised 4 February 2011; accepted for publication 11 February 2011 Some of the morphological characters used in Porifera taxonomy have often been shown to be inconsistent. In the present study, we tested the phylogenetic coherence of currently used taxonomic characters of the calcarean genus Clathrina. For this, 20 species of Clathrina and three other calcinean genera (Ascandra, Guancha, and Leucetta) were sequenced for the ITS and D2 region of the 28S ribosomal DNA. Maximum-likelihood and maximum- parsimony algorithms were used to reconstruct phylogenetic trees. Deep divergences were observed in our tree and Clathrina was shown to be paraphyletic. The major split in our topology showed a clear-cut distinction between sponges with and without tetractine spicules. Moreover, a group of yellow-coloured Clathrina was clearly separated from the remaining white-coloured species. Our results show that the presence of diactines, water-collecting tubes, the degree of cormus anastomosis, and actine shapes do not correlate with the major clades of the calcinean phylogeny. On the other hand, the presence of tripods, the absence of tetractines, and the presence of spines in the apical actine of tetractines seem to be good synapomorphies for clades in our tree. Our results demonstrate that skeleton characters can be reliably used in higher level taxonomy in Clathrinida. © 2011 The Linnean Society of London, Zoological Journal of the Linnean Society, 2011. doi: 10.1111/j.1096-3642.2011.00739.x ADDITIONAL KEYWORDS: Calcarea – Calcinea – Clathrina – molecular phylogenetics – morphological evolution – Porifera. INTRODUCTION typological approach is still applied by research groups around the world. The reason is simple: con- A phylogenetic tree illustrates the flow that deter- sistent phylogenies are scarce for many groups of mines shared homologous characteristics within organisms (Manuel et al., 2003; Dohrmann et al., groups of organisms. More than a century and a half 2006; Pick et al., 2010) and the typological approach ago, the publication of On the Origin of Species set is the method of choice to determine the taxonomic the stage for the idea of a phylogenetically sound position of research specimens. Invertebrate animal classification (Darwin, 1859). Currently, there is groups with simple morphology and a meagre fossil an increasing trend toward using phylogenetics to record fit well into this category. understand the diversity of life, but the traditional For those groups, molecular data seem to be the key to reliable phylogenetic reconstructions, as evolution at the molecular level may be accurately described *Corresponding author. E-mail: [email protected] by statistical models, and topology inference with © 2011 The Linnean Society of London, Zoological Journal of the Linnean Society,2011 1 2 A. L. ROSSI ET AL. molecular data becomes a statistical issue (Nei & Solé-Cava et al., 1991 (2); C. cerebrum (Haeckel, 1872) Kumar, 2000). Once a well-supported molecular phy- (3); C. clathrus (Schmidt, 1864) (3); C. conifera logeny is available, however, dubious morphological Klautau & Borojevic, 2001 (2); C. contorta Minchin, characters should be checked and the taxonomy 1905 (1); C. corallicola Rapp, 2006 (1); C. coriacea re-evaluated (e.g. Waeschenbach et al., 2009; Cárde- (Montagu, 1818) (1); C. cylindractina Klautau, nas et al., 2010). In the case of morphologically simple Solé-Cava & Borojevic, 1994 (4); C. fjordica Azevedo organisms, such as sponges, the problem becomes et al., 2009 (2); C. helveola Wörheide & Hooper, critical (Borchiellini et al., 2000). Porifera systematics 1999 (1); C. luteoculcitella Wörheide & Hooper, 1999 is mainly based on the spicule skeleton, but morpho- (1); C. nanseni (Breitfuss, 1896) (2); C. reticulum logical characters in many sponge groups have often (Schmidt, 1862) (3); C. tetractina Klautau & been shown to be plastic (e.g. Klautau et al., 1999; Borojevic, 2001 (1); and C. wistariensis Wörheide & Bell, Barnes & Turner, 2002; Valderrama et al., 2009) Hooper, 1999 (1). or simply absent (Lazoski et al., 2001). Based on the current taxonomy, Clathrina is dis- In the cosmopolitan genus Clathrina, for instance, tinguishable from closely related genera by character species diagnoses are mainly based on absent or states that might vary even within a single specimen dubious characters (Klautau & Valentine, 2003). Pro- (Klautau & Valentine, 2003). Hence, we have added gressively complex body organizations, however, another six calcinean species from three different are clearly recognized among species of the genus. genera to test the monophyletic condition of the genus Simpler forms have loosely anastomosed tubes with (the number of specimens sequenced is shown in several oscula, whereas more complex forms present parentheses): Ascandra falcata Haeckel, 1872 (2) tightly anastomosed tubes with a sharply distinct (type species); Guancha sp. (3); G. lacunosa Johnston, cormus and fewer oscula with water-collecting tubes 1842 (2); G. ramosa Azevedo et al. 2009 (1); Leucetta (Klautau & Valentine, 2003). In addition, as species in chagosensis Dendy, 1913 (1); and L. microraphis the genus Clathrina exhibit several combinations of Haeckel, 1872 (1). spicule types, these sponges constitute an excellent model system to test the evolution of skeleton and DNA SEQUENCING body organization using a molecular evolutionary The internal transcribed spacer (ITS) and the D2 approach. Thus, this study aims to test the phyloge- region of the 28S ribosomal DNA were sequenced netic coherence of relevant characters in the tax- for our phylogenetic analyses. Multi-copy ribosomal onomy of the calcarean genus Clathrina. genes and spacers may be unreliable in low-level molecular taxonomy studies (Nei & Kumar, 2000). Nonetheless, in calcareous sponges, gene conversion seems to be effective (Wörheide, Nichols & Goldberg, MATERIAL AND METHODS 2004), eliminating paralogy-related problems using TAXA multicopy markers in the group. The class Calcarea is currently divided into two sub- In spite of the extensive experience of our group in classes, Calcinea and Calcaronea. Both are mono- the study of sponge genetics, we were unsuccessful phyletic, according to extensive morphological and when trying to amplify fragments of mitochondrial molecular data (e.g. Manuel et al., 2003; Dohrmann and other nuclear protein-coding genes. Sequencing et al., 2006). Due to a current dispute on the status of mitochondrial DNA of calcareous sponges is not a most Calcinean genera and families (Dohrmann et al., trivial task, and Calcarea remains one of the few 2006) and considering that Clathrina belongs to the metazoan classes for which no mitochondrial genome Calcinea subclass, we have selected two calcaronean has been released, despite several attempts (D. species, Paraleucilla magna Klautau, Monteiro & Lavrov, pers. comm.). Borojevic, 2004 and Sycettusa tenuis Borojevic & Genomic DNA was extracted from ethanol or liquid Klautau, 2000, as outgroups. nitrogen-preserved specimens with the guanidine/ Species names, collection sites, voucher numbers, phenol chloroform protocol (Lôbo-Hajdu et al., 2004) and GenBank accession numbers for the newly gen- or using a Viogene kit and following the manufactur- erated sequences and for those downloaded from er’s instructions. The entire region comprising the GenBank are listed in Table 1. Twenty species of two spacers (ITS1 and ITS2) and the 5.8S ribosomal Clathrina were sequenced for our analysis (the DNA was amplified by PCR with primers anchored on number of specimens sequenced for each species is 18S (5′TCATTTAGAGGAAGTAAAAGTCG3′) and 28S shown in parentheses): Clathrina sp. 1 (3); Clathrina (5′GTTAGTTTCTTTTCCTCCGCTT3′) (Lôbo-Hajdu sp. 2 (1); C. antofagastensis Azevedo et al., 2009 (1); et al., 2004). C. aspina Klautau, Solé-Cava & Borojevic, 1994 (3); The D2 region of the 28S rDNA was amplified C. aurea Solé-Cava et al., 1991 (3); C. brasiliensis by PCR with primers C2F (5′GAAAAGAACTTTG © 2011 The Linnean Society of London, Zoological Journal of the Linnean Society, 2011 PHYLOGENETIC SIGNAL IN SPONGE MORPHOLOGY 3 Table 1. Specimens used in this study with collection sites, voucher numbers, and GenBank accession numbers of DNA sequences GenBank accession number Species Collection site Voucher number ITS 28S Ascandra falcata Mediterranean Sea UFRJPOR 5856 HQ588962 HQ589006 Ascandra falcata Mediterranean Sea UFRJPOR 6320 HQ588963 – Clathrina antofagastensis Chile MNRJ 9289 HQ588985 HQ589003 Clathrina aff. aspina Brazil UFRJPor 5211 HQ588969 – Clathrina aff. aspina Brazil UFRJPor 5245 HQ588998 – Clathrina aff. aspina Brazil URFJPor 5495 – HQ589017 Clathrina aurea Brazil MNRJ 8998 HQ588968 HQ589005 Clathrina aurea Brazil MNRJ 8990 HQ588958 – Clathrina aurea Brazil MNRJ 5170 HQ588960 – Clathrina brasiliensis Brazil UFRJPor 5214 HQ588978 HQ589015